Virtual Motion Detection Method Using Fuzzy Logic for an Embedded System

Autor: C. A. Bonilla Barragán, M.A. Gurrola Navarro, A. S. Medina Vazquez, I. Orduna
Rok vydání: 2019
Předmět:
Zdroj: 2019 IEEE International Conference on Engineering Veracruz (ICEV).
DOI: 10.1109/icev.2019.8920710
Popis: This document presents a method that emulates virtual motion detection with non-invasive sensors for actuator control by using fuzzy logic, to determine the position of an object in three-dimensional space, as well as the direction of its movement. The design and analysis of the fuzzy algorithm were performed with MATLAB using fuzzy logic and by the method of implication Mamdani. An algorithm was generated for architecture with two-dimensional meshes of 10 ×10 sensors and actuators, the algorithm allows to modify the number of linguistic variables, however, for the results presented here, five values were considered. The process was divided into two stages, the first one evaluates individually the distance that each sensor detects, and the second one discriminates different detection areas according to the number of sensors activated. Finally, the viability of its possible implementation in an FPGA was suggested, which would carry out the corresponding operations of the algorithm to generate different levels of amplification in the actuators. The results were analyzed employing test cases to generate the corresponding graphs.
Databáze: OpenAIRE